Patrícia C MastroianniI Ana Regina NotoII José Carlos F GaldurózII
I Departamento de Fármacos e Medicamentos. Faculdade de Ciências Farmacêuticas. Universidade Estadual Paulista Júlio Mesquita. Araraquara, SP, Brasil
II Departamento de Psicobiologia. Escola Paulista de Medicina. Universidade Federal de São Paulo. São Paulo, SP, Brasil Correspondence:
Patrícia C Mastroianni
Faculdade de Ciências Farmacêuticas da UNESP
Rodovia Araraquara-Jaú km 1 14801-902 Araraquara, SP, Brasil E-mail: pmastro@fcfar.unesp.br
Received: 7/23/2007 Reviewed: 11/15/2007 Approved: 1/8/2008
Psychoactive drug advertising:
analysis of scientifi c
information
ABSTRACT
OBJECTIVE: According to the World Health Organization, medicinal drug promotion should be reliable, accurate, truthful, informative, balanced, up-to-date and capable of substantiation. The objective of the present study was to review psychoactive drug advertisements to physicians as for information consistency with the related references and accessibility of the cited references.
METHODS: Data was collected in the city of Araraquara, Southeastern Brazil, in 2005. There were collected and reviewed 152 drug advertisements, a total of 304 references. References were requested directly from pharmaceutical companies’ customer services and searched in UNESP (Ibict, Athenas) and BIREME (SciELO, PubMed, free-access indexed journals) library network and CAPES journals. Advertisement statements were checked against references using content analysis.
RESULTS: Of all references cited in the advertisements studied, 66.7% were accessed. Of 639 promotional statements identifi ed, 346 (54%) were analyzed. The analysis showed that 67.7% of promotional statements in the advertisements were consistent with their references, while the remaining was either partially consistent or inconsistent. Of the material analyzed, an average 2.5 (1–28) references was cited per advertisement. In the text body, there were identifi ed 639 pieces of information clearly associated with at least one cited reference (average 3.5 pieces of information per advertisement).
CONCLUSIONS: The study results evidenced difficult access to the references. Messages on effi cacy, safety and cost, among others, are not always supported by scientifi c studies. There is a need for regulation changes and effective monitoring of drug promotional materials.
DESCRIPTORS: Psychotropic Drugs. Drug Promoter. Products Publicity Control. Medication Systems, Hospital. National Drug Policy. Review [Publication Type].
INTRODUCTION
According to the World Health Organization (WHO), medicinal drug promo-tion should be reliable, accurate, truthful, informative, balanced, up-to-date and capable of substantiation. Text and illustration contents should be consistent with scientifi c information.20
Non-ethical medicinal drug promotion is a major issue worldwide leading to irrational drug use, overprescription, self-medication and drug abuse.1,10,11 This
of Brazilian advertisements are more subjective and incomplete compared to same drug materials advertised in the UK and US.3,15,16
Several studies1,7,12 and systematic reviews19,24 have
described the effect of drug promotional materials on medical prescription. Above all, it is a more serious is-sue for psychoactive drug advertising since it tends to be less informative than other medicinal drug promotional materials.4,9 In addition, these advertisements usually
reinforce stereotypes of the association between gender and psychiatric conditions, showing a disproportion between women and men.14,15
Studies have also reported both misinformation and unbalanced information in drug promotion support-ing their use: indication, presentation and dosage are often included and highlighted by larger font sizes and colorful text, whereas information restraining drug use, such as contraindications, warnings, precautions and adverse reactions, when available, are less evident and not as visible.15,17,22 However, these studies have
not investigated the consistency between promotional claims and related cited references.
Few studies5,23 have addressed reliability, accuracy
and truthfulness of information in drug promotional materials and their related cited references. Although it is a major issue and a requirement for complying with international recommendations,20 national drug
regula-tions (Ministerial Decree No. 3.916/98a), and current
health law (Collegiate Board of Directors’ Resolution [RDC] No. 102/00b), there is no study to date
address-ing it in Brazil.
The purpose of the present study was to review psychoactive drug promotional materials targeted to physicians as for consistency of information with their related references and to assess accessibility of the cited references.
METHODS
The study sample consisted of psychoactive drug advertising materials to physicians provided by phar-maceutical promoters in clinics, hospitals and health units. Data was collected in the city of Araraquara, Southeastern Brazil, in 2005.
Psychoactive drugs are prescription only medicines subject to special control and prescription retention. They can only be advertised to prescribers and dis-pensers (RDC 102/00b, Art. 5a; Art. 13 and HSS/MoH
Administrative Rule No. 344/98c - Art. 12, item II;
a Brasil. Portaria n° 3.916, de 30 de outubro de 1998. Aprova a Política Nacional de Medicamentos. Diário Ofi cial da União. 10 nov 1998. b Brasil. Resolução RDC nº 102, de 30 de novembro de 2000. Aprova o Regulamento sobre propagandas, mensagens publicitárias e promocionais e outras práticas cujo objeto seja a divulgação, promoção ou comercialização de medicamentos de produção nacional ou importados, quaisquer que sejam as formas e meios de sua veiculação, incluindo as transmitidas no decorrer da programação normal das emissoras de rádio e televisão. Diário ofi cial da União. 1 dez 2000.
Art. 16) through promotional materials both delivered at health settings, and published in medical journals and books.
Pharmaceutical companies disseminate the same drug advertising materials nationwide. Thus, materials collected in the city of Araraquara can be deemed representative of psychoactive drug advertisements distributed to prescribers nationwide.
For data collection, physicians in hospitals, clinics and health units were contacted and the purpose of the study was explained. Those who agreed to participate were asked to voluntary retain all promotional materials for the team’s monthly collection.
No differences were seen in materials collected by site of collection since the same promoter of each pharma-ceutical company used to visit the hospitals, clinics and health units and provide the same materials.
References were requested from physician and phar-maceutical customer services (PPCS) through the number provided in the promotional materials. The references made available by PPCS were sent through mail and searched in UNESP (Ibict, Athenas), and BI-REME libraries (Scielo, Pubmed, free-access indexed journals) and other on-line databases available, such as CAPES journals.
Reference information on access source (laboratory, library or unknown) and review status (reviewed, not reviewed and reason for no review) were entered into a database.
Content analysis of each reference obtained followed a guide developed based on a “fl oating reading” of promotional materials and their related references.2
The guide included the following:
Part I – Promotional material information: ID number, drug name, active component, class of therapeutic agent, pharmaceutical company name, PPCS number; cited references and number of references cited;
Part II – Review of cited references: type of re-ference (published and indexed; published and non-indexed; unpublished article; book citation; pharmaceutical material; meeting presentation; association handbook; drug price guide); study category (meta-analysis; clinical trial; cohort stu-dy; case-control stustu-dy; editorial; review; survey; pharmacovigilance study), and consistency level (consistent; inconsistentor somewhat consistent with promotional claims).
•
To prevent both subjective and differing interpreta-tions, content analysis was conducted in duplicate by independent reviewers. In case of disagreement, a third reviewer would review the content. References were classifi ed according to study category and level of evidence and grade of recommendation.21
“Consistent” information was defi ned when promotion-al claims were found in the cited reference; “somewhat consistent” information was defi ned when at least one claim or part of it was found in the cited reference; and “inconsistent” information was defi ned when promo-tional claims were not found in the cited reference.
A database was electronically created in a fl ow chart to manage data collected.
The study was approved by UNIFESP Research Ethics Committee (Protocol No.176/06).
RESULTS
There were collected 167 different psychoactive drug advertising materials from 25 pharmaceutical compa-nies. Most of them promoted antidepressive (41.9%) and anxiolytic drugs (24.5%). Fifteen promotional materials (from four pharmaceutical companies) were excluded as they did not provide any references.
Of 152 promotional materials reviewed, there were on average 2.5 (1–28) references cited per piece, making a total of 395 references. In their body, there were identifi ed 639 pieces of information that were clearly associated to at least one reference cited (mean 3.5 pieces of information per material) (Figure).
Two pharmaceutical companies refused to provide the requested references. Although most companies claimed they did not have any references in their fi les, all those who agreed to collaborate said they will ask their headquarters or medical libraries to provide the requested references. It took three days to six months for companies to send them. Only one pharmaceutical company provided all references requested; the remain-ing provided on average less than 50% of references cited and 107 (27%) references were directly obtained. The remaining 156 (39.5) references were accessed in libraries or other database referred in Methods. A total of 263 were obtained. The remaining references (132; 33.5%) were either not available in the compa-nies’ headquarters or were not indexed in the database searched or unpublished (Table 1).
Most references were from studies published in indexed (291; 73.7%) or non-indexed (13; 3.3%) journals. Based on these 304 references, 260 studies were categorized through accessing their abstracts or entire content.
It was found that the majority of studies were clinical trials (N=116), meaning the highest level of evidence for therapeutic choice, followed by literature reviews (N=94), meaning the lowest level of evidence for therapeutic choice.21 In addition, there were found
non-recommended studies such as pre-clinical studies (N=12), epidemiological surveys (N=6) and life quality assessment (N=1) (Table 2).
In regard to the level of consistency between promo-tional claims and the related cited references, of 263 references available (107 provided by pharmaceuti-cal companies and 156 obtained from libraries and database), 346 claims were reviewed, 54% (346/639) of all claims identifi ed in the advertising materials collected.
Table 1. Reasons for diffi cult access to cited references in psychoactive drug advertising materials. Araraquara, Sou-theastern Brazil, 2005.
Reason N %
Articles not available or not provided by physician and pharmaceutical customer services
52 39.4
Drug price guides 22 16.7
Posters or presentations in
scientifi c meetings 15 11.4
Books 10 7.6
References not cited in
promotional materials 9 6.8
Unpublished pharmaceutical material 7 5.3
Unpublished sales report 5 3.8
Incomplete/inaccurate references 5 3.8
Articles written in French 3 2.3
Articles written in German 3 2.3
Brazilian government offi cial newspaper 1 0.6
Total 132 100
Table 2. Study categories of studies referenced in psycho-active drug advertising materials. Araraquara, Southeastern Brazil, 2005.
Study category N %
Recommended
Clinical trial 116 44.6
Meta-analysis 8 3.1
Cohort study 8 3.1
Pharmacovigilance/case study 9 3.5
Reviews/Guides 94 36.1
Editorial 6 2.3
Not recommended
In vivo/in vitro pre-clinical study 12 4.6
Epidemiological survey 6 2.3
Quality of life assessment 1 0.4
In most claims reviewed (234; 67.7%), sentences or information were identifi ed in the cited reference. In 15.6% (54), information was either incomplete or partly referred to the reference cited; and in 16.7% (58), no information was found in the reference cited.
Inconsistent or somewhat consistent content was due to missing or confl icting information found in the ref-erenced studies, i.e., different drugs studied, reviews of classes of therapeutic agents that did not provide specifi c information on the promoted drug, and different population studied, e.g., young adults when it claimed to be “(…) safe for the elderly”. There were also found pieces of information including extrapolated results with increased level of statistical signifi cance; animal studies but claims were for humans; studies in patients with a single condition and claims about the drug effi cacy for patients with two or more conditions or even about the effi cacy in patients with co-morbidities (e.g., renal or liver failure) not investigated in the study (Table 3).
DISCUSSION
The present study showed diffi cult access to the refer-ences cited in psychoactive drug advertising materials. Villanueva et al23 (2003) reported similar fi ndings after
studying drug advertisements published in six Spanish newspapers. These authors did not have access to 18% of the cited references since they were monographs or unpublished data.23
Prescribers would fi nd even more diffi cult to have access to referenced information in drug promotional materials.
The present study also evidenced a need for changes in current health laws regarding drug promotion regula-tions, banning citation of non-scientifi c based reference, such as drug price guides. Moreover, pharmaceutical companies should be required to make available full original studies in their websites and to cite in promo-tional material studies designed at the recommended level of evidence for therapeutic choice, preferably published studies rather than posters and presentations given at scientifi c meetings.
Promotional materials often include incomplete, sum-marized, inconsistent and different information than that in the referred studies,8 favorably supporting the
thera-peutic indication, drug effi cacy, safety and cost.18,19,24
In contrast to previous studies that have reported low citation of references in advertising claims,15,17,22 in the
present study, 91% of the materials reviewed had at least one reference. It can be noted that today pharmaceuti-cal companies tend to provide references as marketing strategies of drug promotion.23
Pharmaceutical companies frequently draw on random-ized clinical trial studies (44.6%) published in renowned indexed medical journals,5,23 especially intended for
therapeutic choice.21 On the other hand, 36.1% of
refer-ences here studied were literature reviews, which are less recommended21 and in general promotional claims
favorably supported drug use and omitted warnings and precautions to specifi c populations stressed in the reviews. Hence, unbalanced information is usually found in advertising claims.15,17,23
A good number of claims reviewed (67.7%) were consistent with the references because consistency was ascertained when the cited information was identifi ed regardless of any other considerations. RDC 102/00, Art. 15, establishes that “citations, tables or other il-lustrations obtained from scientifi c publications and included in any advertisements or promotional materi-Figure. Flow chart of cited references in psychoactive drug
advertising materials reviewed. Araraquara, Southeastern Brazil, 2005.
15 advertising materials without any references from 4 pharmaceutical
companies
167 psychoactive drug advertising materials
from 25 pharmaceutical companies
152 advertising materials with references from 21 pharmaceutical companies
395 references cited in the materials
639 pieces of information in the advertising materials
with references
Table 3. Reasons for inconsistency or partial consistency be-tween claims in psychoactive drug advertising materials and related references. Araraquara, Southeastern Brazil, 2005.
Reason* N %
Missing information 87 62
Extrapolation regarding indication/
conditions/class of therapeutic agent 12 12
Confl icting information or misinformation 14 10
Extrapolation to humans 8 5.6
Extrapolation of statistical data 7 5
Different population studied 6 4
Extrapolation regarding quality of life 2 1.4
Total 141 100
als have to be accurately reproduced and cite the full reference”. Most citations were from literature reviews that contemplate pros and cons; pros were cited in the promotional materials but only positive results of clini-cal trials were cited.
Studies have reported publication biases,6,13 i.e., studies
with positive results supporting drug use are promptly published while unfavorable data require much more time to be published or even remain unpublished, which could explain the fact that most claims reviewed were consistent with the cited references. Also, because pharmaceutical companies fund clinical trials, the results are reported as part of their marketing strate-gies23 so promotional claims tend to favorably support
drug use.25
However, it is not known whether the same consistency would have been found if all 132 references not avail-able, accounting for 293 claims that were not reviewed (46%), were accessed.
Inaccurate psychoactive drug advertising was found in 32.3% of information reviewed. A similar fi nding was reported by Villanueva et al23 (2003) while studying
promotional claims in antihypertensive drug advertise-ments published in Spanish medical journals and related references (44.1%) and by Gómez-Garcia et al5 (2005)
while reviewing promotional materials provided to primary care physicians in Spain (44.5%).
Ziegler et al25 (1996) studied the accuracy of drug
information provided to physicians by promoters and found that 11% were confl icting. In present study, 10% of inconsistent claims were due to misinformation or confl icting information.
The reasons for inconsistent or somewhat consistent information found in the present study were similar to those reported in other studies in Spain:5,23 missing
information, other drugs studied, patients with condi-tions not investigated in the studies, statistical differ-ences were not signifi cant, extrapolation of indications and conditions and conclusions other than those claimed in the materials. According to RDC 102/00,a it is thus
characterized misleading advertising due to information omission or misinformation.
Misinformation was often found in promotional materi-als claiming effi cacy, safety and low costs. Low-cost claims were referenced to drug pride guides. Although cost-effective claims are usually seen in drug promo-tion,18 there are required cost-effectiveness analyses
comparing treatment costs rather than drug prices.
Health providers have to be more careful and apply their knowledge and better judgment while appraising promotional materials. Pharmaceutical companies are not concerned whether references are correctly cited, references are actually related to the drug promoted and comparisons are made between same classes of therapeutic agents. Health providers are unlikely to judiciously appraise advertising materials provided to them and thus become more prone to be infl uenced by promotional practices.19
In conclusion, the issue requires further discussion. It is proposed content analyses of drug advertising in under-graduate studies, continuing education and permanent education programs to degree providers to encourage rational drug use and to prevent abusive prescription of psychoactive drugs and promotional practices to infl u-ence health providers’ therapeutic choices.
ACKNOWLEDGMENTS
To UNESP Pharmacy students for their help with data material collection.
1. Avorn J, Chen M, Hartley R. Scientifi c versus commercial sources of infl uences on the prescribing
behavior of physicians. Am J Med. 1982;73(1):4-8.
2. Bardin L. Análise de conteúdo. Lisboa: Edições 70; 1977.
3. Barros JAC. One more case of the double standard: discrepancies between drug information provided to Brazilian and American physicians.
Pharmacoepidemiol Drug Saf. 2000;9(4):281-7.
4. Carlini EA. O uso e a propaganda de medicamentos.
Exemplos com psicotrópicos. Rev Assoc Bras Psiquiatr.
1983;5(18):152-8.
5. Gómez-Garcia B, Navas Cutanda J, Richart Rufi no MJ, Riviera Casares F, Rodríguez Górriz E, Gómez Moruno C. La informacion científi ca que la industria farmacéutica proporciona a los médicos de familia.
Aten Primaria. 2005;36(1):14-8.
6. Easterbrook PJ, Berlin JA, Gopalan R, Matthews
DR. Publication bias in clinical research. Lancet.
1991:337(8746):867-72
7. Fagundes MJD, Soares MGA, Diniz NM, Pires JR, Garrafa V. Análise bioética da propaganda e
publicidade de medicamentos. Cienc Saude Coletiva.
2007;12(1):221-9.
8. Gutknecht DR. Evidence-based advertising? A
survey of four major journals. J Am Board Fam Pract.
2001;14(3):197-200.
9. Hemminki E. The quality of drug advertisements in two Finnish medical journals. A comparison between
psychotropic and other drug advertisements. Soc Sci
Med. 1973;7(1):51-9.
10. Hogerzeil HV. Promoting rational prescribing: an
international perspective. Br J Clin Pharmacol.
1995;39(1):1-6.
11. Kessler DA, Pines WL. The federal regulation of
prescription drug advertisement and promotion. JAMA.
1990;264(18):2409-15.
12. Lexchin J. Interactions between physicians and the pharmaceutical industry: what does the literature say?
CMAJ. 1993;149(10):1401-7.
13. Loannidis JPA. Effect of the statistical Signifi cance of results on the time to completion and publication of
randomized effi cacy trials. JAMA. 1998:279(4):281-6.
14. Lövdahl U, Riska E. The construction of gender and mental health in Nordic psychotropic-drug advertising.
Int J Health Serv. 2000;30(2):387-406.
15. Mastroianni PC, Galduróz JCF, Carlini EA. Infl uence of the legislation on the advertisement of
psychoactive medications in Brazil. Rev Bras Psiquiatr.
2003;25(3):146-55.
16. Mastroianni PC, Galduróz JCF, Carlini EA. Psychoactive drug advertising a comparison of technical information from three countries: Brazil;
United States and United Kingdom. Sao Paulo Med J.
2005;123(5):208-14.
17. Mejia R, Avalos A. Material informativo entregado
por los agentes de propaganda medica. Medicina (B
Aires). 2001;61(3):315-8.
18. Neumann PJ, Zivin-Bambauer K, Ramakrishnan V, Stewart KA. Bell CM. Economic messages in prescription drug advertisements in medical journal.
Med Care. 2002;40(9):840-5.
19. Norris P, Herxheimer A, Lexchin J, Mansfi eld P. Drug Promotion. What we know, what we have yet to learn. Ginebra; World Health Organization; 2005.
20. World Health Organization. Criterios éticos para la promoción de medicamentos. Ginebra; 1988.
21. Oxman AD, Sackett DL, Guyatt GH. Users Guide to the Medical literature: I how to get started. The
Evidence-Based Medicine Working Group.JAMA 1993;
270: 2093-2095
22. Pizzol FD, Silva T, Schenkel EP. Análise da adequação das propagandas de medicamentos dirigidas à
categoria médica distribuídas no Sul do Brasil. Cad
Saude Publica. 1998;14(1):85-91.
23. Villanueva P, Peiró S, Librero J, Pereiró I. Accuracy of pharmaceutical advertisement in medical journals.
Lancet. 2003;361(9351):27-32.
24. Warzana A. Physicians and the pharmaceutical industry: is a gift ever just a gift? JAMA.
2002;283(3):373-80.
25. Ziegler MG, Lew P, Singer BC. The accuracy of Drug Information from pharmaceutical sales representatives.
JAMA. 1996;273(16):1296-8.
REFERENCES
Presented at the Simpósio de Propaganda de Medicamentos, held during the 53th Jornada Farmacêutica in Aug 20th 2006, Araraquara, Brazil and at the 3rd Reunião Ordinária da Câmara Setorial de Propaganda e Publicidade de Produtos sujeitos à Vigilância Sanitária held in Oct 4th 2006 in Brasília, Brazil.
